Tradeoff Between Spectrum Cost and Quality of Service in a Cognitive Radio Network

This paper studies how to manage a cognitive radio network (CRN) by trading off the cost of spectrum acquisition and the quality of service (QoS) in serving secondary users. Due to bursty nature of primary users' traffic and random realization of spectrum holes, it is difficult for a CRN operator to guarantee the QoS to its users by using the cheap but unreliable spectrum sensing only. Once the sensed available spectrum turns out to be not enough to satisfy users' QoS requirement, the CRN operator needs to lease additional spectrum from licensed network operators, which will incur high leasing cost and threatens the long-run viability of the CRN network. In this paper, we propose a real-time decision-making algorithm for the CRN operator to acquire spectrum based on a Lyapunov optimization framework. We successfully minimize the total sensing and leasing cost, while guaranteeing each user's average queuing delay below a certain designed threshold